Sage 3.3 sources are out

[mabshoff]

Hello folks,

the final 3.3 sources are out and now available at

http://www.sagemath.org/src/

as well as some of the mirrors already (at least Seattle II, Boston,
Germany and France since I personally pushed or pulled the update).
Over 3.3.rc3 there were as expected very few merges:

Merged in Sage 3.3.final:

#1641: William Stein: Make ATLAS restart build on tolerance error
[Reviewed by Michael Abshoff]
#5329: William Stein: sage-3.3.rc3 -- latex mistake [Reviewed by
Michael Abshoff]
#5327: Robert Miller: multiple edge plots use symbolic computations
[Reviewed William Stein]

Expect an official announcement in the next couple hours as I finish
the release notes.

Due to the speedy 3.4 release which seems to be happening as planned
(thank everybody who did a ReST related review so far: Carl Witty
(mutliple), John Palmieri (multiple), David Joyner - I hope I am not
forgetting anyone) we do not plan to make *any* binaries available,
but if there is sufficient demand we might change our minds. *If* we
do binaries we need to fix two distribution related tickets to make
the binaries work well. They were blockers against 3.3, but once we
agreed that we wouldn't do binaries I bumped them to 3.4 so that 3.3
could get out faster :). Since the Sage builds have already finished
all that is required would be to apply two patches to each build and
then bbist, so this is doable quickly.

Cheers,

Michael

[Jan Groenewald]

Hi

On Sat, Feb 21, 2009 at 10:39:52PM -0800, mabshoff wrote:
> the final 3.3 sources are out and now available at
> http://www.sagemath.org/src/

On a P4 with Ubuntu 8.04.2 with gcc 4.2.3-1ubuntu6 and
texlive-common 2007-13. I installed the biopython
and database* and sagetex and jsmath-fonts optional
packages.

As I did make html for the docs in sage-devel/doc,
I got this error:
http://users.aims.ac.za/~jan/make_doc_error.log

As I did make test, there was one error:
http://users.aims.ac.za/~jan/test.log
( sage -t "devel/sage/sage/schemes/elliptic_curves/ell_rational_field.py" )

regards,
Jan

--
.~.
/V\ Jan Groenewald
/( )\ www.aims.ac.za
^^-^^

[mabshoff]

On Feb 22, 11:16 pm, Jan Groenewald <j...@aims.ac.za> wrote:
> Hi

Hi Jan,

> On Sat, Feb 21, 2009 at 10:39:52PM -0800, mabshoff wrote:
> > the final 3.3 sources are out and now available at
> >    http://www.sagemath.org/src/
>
> On a P4 with Ubuntu 8.04.2 with gcc 4.2.3-1ubuntu6 and
> texlive-common 2007-13. I installed the biopython
> and database* and sagetex and jsmath-fonts optional
> packages.
>
> As I did make html for the docs in sage-devel/doc,
> I got this error:http://users.aims.ac.za/~jan/make_doc_error.log

My guess would be you need to install amsmath and potentially a couple
other TeX packages. Since the LaTeX based documentation goes away in
Sage 3.4 I don't think this is worth opening a ticket. A large part of
the reason to switch to ReST was that we will no longer require LaTeX
for the default documentation build since making it to work was always
a pain.

> As I did make test, there was one error:http://users.aims.ac.za/~jan/test.log
> ( sage -t  "devel/sage/sage/schemes/elliptic_curves/ell_rational_field.py" )

Oh, the irony :). One of the installed optional database spkgs (not
sure which name it has, but it is the one providing data for elliptic
curves - it is by John Cremona IIRC) make a failure we check for in
the doctests in the default config into an operation that works. I
have made this #5346. Thanks for reporting the failure.

> regards,
> Jan

Cheers,

Michael

[Jan Groenewald]

Hi

On Sun, Feb 22, 2009 at 11:39:47PM -0800, mabshoff wrote:
> > As I did make html for the docs in sage-devel/doc,
> > I got this error:http://users.aims.ac.za/~jan/make_doc_error.log
>
> My guess would be you need to install amsmath and potentially a couple
> other TeX packages. Since the LaTeX based documentation goes away in
> Sage 3.4 I don't think this is worth opening a ticket. A large part of
> the reason to switch to ReST was that we will no longer require LaTeX
> for the default documentation build since making it to work was always
> a pain.

OK, in that case I won't even bother. ams-math, texlive-recommended,
texlive-all, and a host of other packages are installed though.

> > As I did make test, there was one error:http://users.aims.ac.za/~jan/test.log
> > ( sage -t  "devel/sage/sage/schemes/elliptic_curves/ell_rational_field.py" )
>
> Oh, the irony :). One of the installed optional database spkgs (not
> sure which name it has, but it is the one providing data for elliptic
> curves - it is by John Cremona IIRC) make a failure we check for in
> the doctests in the default config into an operation that works. I
> have made this #5346. Thanks for reporting the failure.

database_cremona_ellcurve-20071019.spkg, I guess.

PS. I will now be permanently patching user_conf.py to set autosave_interval higher
from 3*60 to 120*60 and and max_history_length lower from 500 to 50. It would be
great if the defaults here were reconsidered.

regards,
Jan

[mabshoff]

On Feb 23, 12:00 am, Jan Groenewald <j...@aims.ac.za> wrote:
> Hi

Hi Jan,

> On Sun, Feb 22, 2009 at 11:39:47PM -0800, mabshoff wrote:
> > > As I did make html for the docs in sage-devel/doc,
> > > I got this error:http://users.aims.ac.za/~jan/make_doc_error.log
>
> > My guess would be you need to install amsmath and potentially a couple
> > other TeX packages. Since the LaTeX based documentation goes away in
> > Sage 3.4 I don't think this is worth opening a ticket. A large part of
> > the reason to switch to ReST was that we will no longer require LaTeX
> > for the default documentation build since making it to work was always
> > a pain.
>
> OK, in that case I won't even bother. ams-math, texlive-recommended,
> texlive-all, and a host of other packages are installed though.

Strange. The documentation build did work for us on the official test
machine we used to build the documentation. But all I can say is good
riddance to the old documentation system :)

> > > As I did make test, there was one error:http://users.aims.ac.za/~jan/test.log
> > > ( sage -t  "devel/sage/sage/schemes/elliptic_curves/ell_rational_field.py" )
>
> > Oh, the irony :). One of the installed optional database spkgs (not
> > sure which name it has, but it is the one providing data for elliptic
> > curves - it is by John Cremona IIRC) make a failure we check for in
> > the doctests in the default config into an operation that works. I
> > have made this #5346. Thanks for reporting the failure.
>
> database_cremona_ellcurve-20071019.spkg, I guess.

Yeah, that sounds about right. It might have been William who created
the spkg, but I believe those are John's data.

> PS. I will now be permanently patching user_conf.py to set autosave_interval higher
> from 3*60 to 120*60 and and max_history_length lower from 500 to 50. It would be
> great if the defaults here were reconsidered.

We fixed the bug in Sage 3.3 that caused the excessive number of
snapshots to be kept around and now a new snaphots will be kept only
in case that the content of the worksheet changed. It still does the
snapshotting every so often, it just deletes it when no changes are
detected. This is suboptimal since we still do IO and use CPU cycles
and should be improved via another ticket we already opened.
Unfortunately the current code does not clean out old excessive
snapshots, but I believe such a script could be easily written.

> regards,
> Jan

Cheers,

Michael

[John H Palmieri]

On Feb 22, 11:16 pm, Jan Groenewald <j...@aims.ac.za> wrote:

> Hi
>
> On Sat, Feb 21, 2009 at 10:39:52PM -0800, mabshoff wrote:
> > the final 3.3 sources are out and now available at
> >    http://www.sagemath.org/src/
>
> On a P4 with Ubuntu 8.04.2 with gcc 4.2.3-1ubuntu6 and
> texlive-common 2007-13. I installed the biopython
> and database* and sagetex and jsmath-fonts optional
> packages.
>
> As I did make html for the docs in sage-devel/doc,
> I got this error:http://users.aims.ac.za/~jan/make_doc_error.log

I've had this problem many times before, and I'm looking forward to
the end of it with sage 3.4. Anyway, try deleting the file 'sage-
devel/doc/ref/ref.toc'. (I think that's the right path; otherwise,
look for ref.toc and delete it wherever it happens to be.)

Good luck,
John

[kcrisman]

Upgraded from 3.3.rc2 more or less successfully. But:

Upon restarting, got 3.3 banner but then immediate ImportError

ImportError: No module named newforms
Error importing ipy_profile_sage - perhaps you should run %upgrade?
WARNING: Loading of ipy_profile_sage failed.

and then another one, ImportErrorPython 2.5.2

ImportError: No module named completer

**********************************************************************

Oops, IPython crashed. We do our best to make it stable, but...


*Then* I restarted again, and got the following:

----------------------------------------------------------------------
| Sage Version 3.3, Release Date: 2009-02-21 |
| Type notebook() for the GUI, and license() for information. |
----------------------------------------------------------------------
The SAGE install tree may have moved.
Regenerating Python.pyo and .pyc files that hardcode the install PATH
(please wait at
most a few minutes)...
Do not interrupt this.
Warning: something went wrong updating the easy-install.pth file.

But from then on everything seems to be working, starting with
sage: 2+2
Just FYI in case this helps track down something else which you've
been noticing.
- kcrisman

[mabshoff]

On Feb 23, 10:42 am, kcrisman <kcris...@gmail.com> wrote:

Hi,

> Upgraded from 3.3.rc2 more or less successfully.  But:
>
> Upon restarting, got 3.3 banner but then immediate ImportError

<SNIP>

> Just FYI in case this helps track down something else which you've
> been noticing.

I can also reproduce this in a 3.3 binary I bdited myself: This is now
#5351. But I could not cause a failure. Overall this is a serious
problem, so I made it critical for 3.4.

> - kcrisman

Cheers,

Michael

[William Stein]

+1 -- I would be very happy if that change were made officially.
Please submit a patch / ticket.

William

[Stan Schymanski]

I get a similar error message after compiling from scratch on Intel Mac
OSX 10.4.11. Is this related?

See full report below.

Cheers
Stan

----------------------------------------------------------------------
| Sage Version 3.3, Release Date: 2009-02-21 |
| Type notebook() for the GUI, and license() for information. |
----------------------------------------------------------------------

---------------------------------------------------------------------------
ImportError Traceback (most recent call last)

/Users/sschym/Programs/sage-3.3/local/lib/python2.5/site-packages/IPython/ipmaker.pyc
in force_import(modname)
64 reload(sys.modules[modname])
65 else:
---> 66 __import__(modname)
67
68

/Users/sschym/Programs/sage-3.3/local/bin/ipy_profile_sage.py in <module>()
5 preparser(True)
6
----> 7 import sage.all_cmdline
8 sage.all_cmdline._init_cmdline(globals())
9

/Users/sschym/Programs/sage-3.3/local/lib/python2.5/site-packages/sage/all_cmdline.py
in <module>()
12 try:
13
---> 14 from sage.all import *
15 from sage.calculus.predefined import x
16 preparser(on=True)

/Users/sschym/Programs/sage-3.3/local/lib/python2.5/site-packages/sage/all.py
in <module>()
70 get_sigs()
71
---> 72 from sage.rings.all import *
73 from sage.matrix.all import *
74

/Users/sschym/Programs/sage-3.3/local/lib/python2.5/site-packages/sage/rings/all.py
in <module>()
72
73 # Number field
---> 74 from number_field.all import *
75
76 # p-adic field

/Users/sschym/Programs/sage-3.3/local/lib/python2.5/site-packages/sage/rings/number_field/all.py
in <module>()
1 from number_field_base import is_NumberField
2
----> 3 from number_field import (NumberField, CyclotomicField,
QuadraticField,
4 is_CyclotomicField, is_QuadraticField,
5 is_AbsoluteNumberField,

/Users/sschym/Programs/sage-3.3/local/lib/python2.5/site-packages/sage/rings/number_field/number_field.pyc
in <module>()
182 from sage.structure.parent_gens import ParentWithGens
183 import number_field_element
--> 184 import number_field_element_quadratic
185 from number_field_ideal import convert_from_zk_basis,
NumberFieldIdeal, is_NumberFieldIdeal, NumberFieldFractionalIdeal
186 from sage.rings.number_field.number_field_ideal_rel import
NumberFieldFractionalIdeal_rel

ImportError:
dlopen(/Users/sschym/Programs/sage-3.3/local/lib/python2.5/site-packages/sage/rings/number_field/number_field_element_quadratic.so,
2): no suitable image found. Did find:

/Users/sschym/Programs/sage-3.3/local/lib/python2.5/site-packages/sage/rings/number_field/number_field_element_quadratic.so:
file to short

Error importing ipy_profile_sage - perhaps you should run %upgrade?
WARNING: Loading of ipy_profile_sage failed.

[mabshoff]

On Feb 23, 11:22 pm, Stan Schymanski <schym...@gmail.com> wrote:
> I get a similar error message after compiling from scratch on Intel Mac
> OSX 10.4.11. Is this related?

I don't think so, the problem might only apply to upgrade.

> See full report below.
>
> Cheers
> Stan

<SNIP>

The linker indicate a library that is "too short", i.e. like damaged.
Can you post a link to the compressed install.log?

Cheers,

Michael

[Stan Schymanski]

Hi Michael,

Thanks a lot for your quick reply. Before I upload the install.log
somewhere, could the error have been caused by an interrupted make the
first time (ran out of disc space)? I re-run make successfully and then
found the problem when starting sage.

Stan

[Georg S. Weber]

On 24 Feb., 08:22, Stan Schymanski <schym...@gmail.com> wrote:
> I get a similar error message after compiling from scratch on Intel Mac
> OSX 10.4.11. Is this related?
>
> See full report below.
>
> Cheers
> Stan

Hi Stan,

I use OS X 10.4.11 on my Intel Mac (MacBook Core2Duo), too, and I have
not seen these problems with Sage 3.3. The Xcode version I use is the
latest one, Xcode 2.5 --- what do you use? What is the output of "gcc
--version"? (There should be some number "5360" or similar, I don't
remember exactly).

Cheers,
gsw

[Stan Schymanski]

Hi Georg and Michael,

Thanks for your help. I use Xcode 2.4.1 only. I suppose I should upgrade
then? After upgrading, can I run make again or do I have to delete the
directory and start again? Here is my gcc version output:

i686-apple-darwin8-gcc-4.0.1 (GCC) 4.0.1 (Apple Computer, Inc. build 5367)
Copyright (C) 2005 Free Software Foundation, Inc.

I put the install.log here:
ftp://ftp.bgc-jena.mpg.de/pub/outgoing/sschym/install.log.zip

Cheers
Stan

--
________________________________________

Stan Schymanski
Scientist
Max Planck Institute for Biogeochemistry
Postfach 10 01 64
D-07701 Jena

Phone: +49.3641.576264
Fax: +49.3641.577274
WWW: http://www.bgc-jena.mpg.de/~sschym

Biospheric Theory and Modelling Group
http://www.bgc-jena.mpg.de/bgc-theory/
_________________________________________

[mabshoff]

On Feb 24, 12:18 am, Stan Schymanski <schym...@gmail.com> wrote:
> Hi Georg and Michael,

Hi Stan,

> Thanks for your help. I use Xcode 2.4.1 only. I suppose I should upgrade
> then? After upgrading, can I run make again or do I have to delete the
> directory and start again? Here is my gcc version output:
>
> i686-apple-darwin8-gcc-4.0.1 (GCC) 4.0.1 (Apple Computer, Inc. build 5367)
> Copyright (C) 2005 Free Software Foundation, Inc.
>
> I put the install.log here:ftp://ftp.bgc-jena.mpg.de/pub/outgoing/sschym/install.log.zip

Gotta run, so two tricks:

* touch the extension missing, run sage -b

You might have to do that for various extensions.

Ff that doesn't work:

./sage -ba

this will take a while, but will rebuild the library from scratch. If
you ran out of space this would be consistent with what we see.

Note that there is a ticket to check for a minimum amount of space and
refuse to build/upgrade Sage if the available space is too low.

> Cheers
> Stan
>

Cheers,

Michael

[Stan Schymanski]

Hi Michael,

Thanks a lot for your help, it works now (./sage -ba did it for me). For
the record, I did not upgrade to Xcode 2.5, still running Xcode 2.4.1.

Cheers
Stan

[mabshoff]

On Feb 24, 1:37 am, Stan Schymanski <schym...@gmail.com> wrote:
> Hi Michael,

Hi Stan,

> Thanks a lot for your help, it works now (./sage -ba did it for me).

Good, I am not surprised. I should really fix that ticket to check for
a minimum amount of space to avoid this happening to other people
again :). There are some other things I meant to fix anyway.

> For the record, I did not upgrade to Xcode 2.5, still running Xcode 2.4.1.

Xcode 2.4.1 by Apple is a known good and supported compiler to build
Sage on OSX 10.4 and 10.5, so no need to upgrade.

> Cheers
> Stan

Cheers,

Michael

[Marshall Hampton]

I don't think Jan has a trac account. I have created #5371 to address
this problem. Because of the cleaning up of the autosaves in 3.3, I
don't think such a long autosave interval is desirable, so I am
suggesting an interval of 5 minutes and a max history of 50 be the
defaults. I should be able to get a patch in for sage-3.4.1.

-Marshall